benchgap
Moonshot · model

Kimi K2.7 Code benchmark scores

As of 2026-11-21, Kimi K2.7 Code (Moonshot) has measured scores on 21 benchmarks and estimated scores on 124 more.

BenchmarkScoreSource
AA-GPQA Diamond89.6%measured
AA-HLE35.0%measured
AA-SciCode47.8%measured
AIME2695.9%estimated ± 1.2 pp, medium confidence
Artificial Analysis Intelligence Index27.2%estimated ± 7.2 pp, medium confidence
CritPt10.0%measured
GDPval-AA26.9%measured
GPQA Diamond89.8%estimated ± 2.4 pp, high confidence
HLE38.1%estimated ± 8.5 pp, medium confidence
IFBench62.1%estimated ± 6.6 pp, medium confidence
AA-LCR79.3%measured
LiveCodeBench v687.4%estimated ± 2.2 pp, medium confidence
AA-Omniscience Accuracy38.1%estimated ± 7.7 pp, medium confidence
SciCode40.9%estimated ± 2.9 pp, low confidence
Terminal-Bench 2.182.9%estimated ± 6.7 pp, medium confidence
BrowseComp78.6%estimated ± 7.8 pp, medium confidence
HLE w/ tools46.8%estimated ± 5.3 pp, medium confidence
IFEval92.4%estimated ± 2.8 pp, high confidence
LongBench v265.1%estimated ± 5.5 pp, medium confidence
VITA-Bench46.6%estimated ± 9.9 pp, low confidence
τ²-bench results90.1%measured
AA Coding Index60.8%measured
APEX-Agents19.7%estimated ± 9.4 pp, low confidence
APEX-Agents-AA34.6%estimated ± 5.5 pp, medium confidence
BioMysteryBench (human-difficult)43.8%estimated ± 6.8 pp, low confidence
DeepSearchQA85.8%estimated ± 4.5 pp, high confidence
SWE-bench Verified80.9%estimated ± 4.4 pp, high confidence
BFCL v473.8%estimated ± 2.7 pp, high confidence
CyberGym62.8%estimated ± 5.3 pp, low confidence
JobBench48.2%estimated ± 7.9 pp, medium confidence
SkillsBench56.0%estimated ± 9.1 pp, low confidence
SWE-bench Pro54.8%estimated ± 4.5 pp, high confidence
WideResearch71.6%estimated ± 6.7 pp, low confidence
AA Agentic Index22.5%measured
MMLU-Pro84.9%estimated ± 7.0 pp, medium confidence
AA-IFBench63.1%measured
FailureSensorIQ64.3%estimated ± 2.1 pp, high confidence
GPQA90.1%estimated ± 6.1 pp, medium confidence
MMLU88.2%estimated ± 8.3 pp, low confidence
Vals GPQA Diamond88.7%estimated ± 3.2 pp, high confidence
Vals LiveCodeBench82.1%measured
Vals MMLU-Pro87.1%estimated ± 3.0 pp, high confidence
MMMLU89.5%estimated ± 0.7 pp, medium confidence
Chess Puzzles14.0%measured
AA MMLU-Pro88.9%estimated ± 1.3 pp, high confidence
Gert Labs64.1%estimated ± 7.8 pp, low confidence
AA-AnalystAgent46.5%estimated ± 3.7 pp, medium confidence
AA Harvey LAB v1.082.2%estimated ± 1.2 pp, medium confidence
AA Tau3 Banking28.9%estimated ± 3.7 pp, high confidence
AA Terminal-Bench 2.181.9%estimated ± 2.5 pp, high confidence
ApprenticeBench14.9%estimated ± 11.0 pp, low confidence
ARC-AGI-182.2%estimated ± 7.8 pp, low confidence
ARC-AGI-249.8%estimated ± 14.4 pp, low confidence
CursorBench 3.138.6%estimated ± 3.4 pp, low confidence
CursorBench 3.249.7%measured
EBR-bench5.7%estimated ± 12.7 pp, low confidence
Terminal-Bench 3.010.6%estimated ± 6.9 pp, low confidence
FrontierCode 1.1 Main33.0%estimated ± 1.6 pp, low confidence
FrontierMath v2 (Tier 4)12.2%measured
FrontierMath v2 (Tiers 1-3)54.0%measured
FrontierSWE v29.4%estimated ± 9.5 pp, medium confidence
Furniture Assembly25.6%estimated ± 14.1 pp, low confidence
MaxxBench36.6%estimated ± 10.3 pp, low confidence
MazeBench0.4%estimated ± 5.5 pp, low confidence
Mystery Game Puzzles9.0%estimated ± 10.3 pp, low confidence
NYT Connections51.6%estimated ± 12.6 pp, low confidence
OSWorld-Verified69.6%estimated ± 8.4 pp, low confidence
Terminal-Bench 4.12.2%estimated ± 9.2 pp, low confidence
Terminal-Bench Hard44.7%measured
Vals SWE-bench78.2%measured
Terminal-Bench 2.1 (Vals AI)67.0%measured
VulcanBench v381.9%estimated ± 5.6 pp, low confidence
CWE-Bench-AA34.4%estimated ± 10.1 pp, low confidence
AA Cyber Index15.0%estimated ± 9.2 pp, low confidence
DeepsecBench-AA1.9%estimated ± 8.9 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)52.5%estimated ± 7.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)55.1%estimated ± 0.8 pp, low confidence
AA ITBench40.0%estimated ± 4.0 pp, high confidence
Bug Hunt Bench9.1%estimated ± 9.0 pp, low confidence
CursorBench 4.020.6%estimated ± 3.0 pp, medium confidence
CWE-bench v136.4%estimated ± 6.4 pp, low confidence
DeepSWE72.9%estimated ± 3.8 pp, medium confidence
DRACO74.0%estimated ± 2.0 pp, medium confidence
FrontierCode 2.014.9%estimated ± 5.4 pp, low confidence
FrontierPhysics0.0%estimated ± 6.1 pp, low confidence
HLE w/o tools34.5%estimated ± 5.4 pp, medium confidence
OSWorld 2.1 Offline3.7%estimated ± 1.9 pp, low confidence
PostTrainBench v1.122.6%estimated ± 3.7 pp, high confidence
ProgramBench53.6%measured
SpireClimb0.0%estimated ± 10.3 pp, low confidence
StealthMark33.6%estimated ± 9.7 pp, low confidence
Toolathlon-Verified51.5%estimated ± 2.3 pp, low confidence
VulcanBench Frontier v488.5%estimated ± 1.2 pp, medium confidence
GDP.pdf17.0%estimated ± 4.6 pp, medium confidence
AA Terminal-Bench 4.022.3%estimated ± 13.7 pp, low confidence
ARC-AGI-31.3%estimated ± 9.8 pp, medium confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, low confidence
cursorBench4144.8%estimated ± 3.7 pp, medium confidence
MKA22.6%estimated ± 6.4 pp, low confidence
SkillsBench official leaderboard (BenchFlow)43.8%estimated ± 6.2 pp, medium confidence
Vibe Code Bench34.9%estimated ± 10.0 pp, low confidence
FACTS Multimodal34.0%estimated ± 5.6 pp, low confidence
Hermes Bench57.4%estimated ± 2.9 pp, high confidence
Hermes Index24.4%estimated ± 8.7 pp, medium confidence
ExploitBench13.7%estimated ± 13.4 pp, low confidence
ExploitGym1.4%estimated ± 10.6 pp, low confidence
C-Eval93.4%estimated ± 1.2 pp, low confidence
Claw-Eval70.6%estimated ± 6.3 pp, low confidence
DeepPlanning48.2%estimated ± 13.1 pp, low confidence
HMMT Feb 202688.1%estimated ± 1.5 pp, medium confidence
HMMT Nov 202594.7%estimated ± 1.5 pp, medium confidence
MCP Atlas76.0%measured
MMAnswerBench87.6%estimated ± 2.4 pp, medium confidence
MMLU-Redux95.1%estimated ± 3.3 pp, high confidence
NL2Repo44.3%estimated ± 4.7 pp, high confidence
PostTrainBench v122.1%estimated ± 3.6 pp, high confidence
QwenClawBench63.0%estimated ± 2.6 pp, high confidence
Terminal-Bench 2.063.4%estimated ± 5.4 pp, medium confidence
Toolathlon50.2%estimated ± 6.3 pp, medium confidence
AA LiveCodeBench90.6%estimated ± 2.6 pp, medium confidence
HealthBench Hard28.9%estimated ± 8.6 pp, low confidence
React Native Evals79.4%estimated ± 3.2 pp, high confidence
ResearchClawBench18.9%estimated ± 1.8 pp, high confidence
SWE-Rebench58.1%estimated ± 8.9 pp, low confidence
PostTrainBench20.1%estimated ± 3.7 pp, high confidence
DAYJOB0.4%estimated ± 6.3 pp, low confidence
DAYJOB Healthcare0.1%estimated ± 6.1 pp, low confidence
BioMysteryBench (human-solvable)83.3%estimated ± 0.6 pp, low confidence
HealthBench (raw)44.0%estimated ± 6.0 pp, low confidence
HealthBench (length-adjusted)51.7%estimated ± 3.4 pp, low confidence
HealthBench Professional58.1%estimated ± 4.5 pp, low confidence
HealthBench Professional (raw)54.8%estimated ± 4.5 pp, medium confidence
HLE-Verified16.6%estimated ± 2.0 pp, low confidence
LABBench276.4%estimated ± 1.8 pp, low confidence
ScarfBench26.8%estimated ± 9.1 pp, low confidence
BlenderWorld (text-only)70.6%estimated ± 4.0 pp, low confidence
Cookie-Bench75.2%estimated ± 2.5 pp, low confidence
SimpleQA33.4%estimated ± 7.8 pp, low confidence
OpenHarmony Bench v1.052.1%measured
SEC-Bench Pro22.4%estimated ± 12.3 pp, low confidence
MRCRv285.7%estimated ± 11.3 pp, low confidence
MMLongBench-Doc61.4%estimated ± 2.6 pp, low confidence
SWE-Marathon53.9%estimated ± 12.9 pp, low confidence
BioBasicsBench33.9%estimated ± 2.8 pp, low confidence
ExploitBench Internal Port0.0%estimated ± 1.4 pp, low confidence